Rémi Gribonval

Directeur de Recherche / Senior Research Scientist, INRIA

Centre de Recherche INRIA Rennes - Bretagne Atlantique, Campus de Beaulieu, F-35042 Rennes cedex, France. E-mail: Firstname.Lastname@inria.fr Tel.: (+33/0) 299 842 506 Fax: (+33/0) 299 847 171

NEWS

• LVA/ICA 2010, the 9th International Conference on Latent Variable Analysis and Signal Separation (formerly the International Conference on Independent Component Analysis and Signal Separation) will be held in Saint-Malo, France, September 27-30 2010. Here is the call for papers
• Current internship offers
• Proposition de stage: Localisation de sources dans le cerveau
• Proposition de stage: Techniques issues des bases de données pour la décomposition atomique rapide de signaux sonores
• Proposition de stage: Aspects statistiques de l'apprentissage de dictionnaire en traitement du signal
• Proposition de stage: Apprentissage automatique et parcimonie

SHORT CV

I graduated from Ecole Normale Supérieure, Paris, in 1997. I prepared my Ph.D. with S.Mallat jointly at the Centre de Mathématiques Appliquées of Ecole Polytechnique and at IRCAM, Paris. I obtained my Ph.D. degree in applied mathematics from the Université Paris-IX Dauphine in 1999. In 1999-2000 I visited the Industrial Mathematics Institute at the University of South Carolina as a postdoc in the Wavelet Ideal Data Representation NSF/KDI research initiative. In 2000 I joined INRIA as Chargé de recherche at IRISA, Rennes in the SIGMA2 project. Since 2001 I am affiliated with the METISS project. I obtained my Habilitation à Diriger des Recherches (HDR) in Mathematics from the Université de Rennes I in 2007. Since 2009 I hold a position of Directeur de Recherche (Senior Research Scientist) with INRIA.

MAIN RESEARCH INTERESTS

• Mathematical aspects of signal processing, machine learning and pattern recognition
  
• Theory and algorithms for sparse representations and nonlinear approximations with redundant dictionaries
• Audio signal processing
• Source localization and separation
  
• Time-frequency and time-scale methods

RESEARCH PROJECTS

• ECHANGE project (2009-2011), funded by the French ANR, on Compressed Sensing of Acoustic Fields
• SMALL project (2009-2011), funded by the EU FP7 FET-Open program, on Sparse Models, Algorithms and Learning for Large-Scale Data
• INRIA-EPFL « Équipe associée »  (2007-2009).  SPARS: efficient sparse representations for high dimensional signal processing
  

Ph.D. STUDENTS

• Ngoc Duong  Source localization in reverberant environments    (since 2008)
  
• Prasad Sudhakar  Sparse models for the reconstruction of incomplete audio data  (since 2007)
  
• Boris Mailhé,    2009, Joint sparse modeling of multichannel signals.
  
• Simon Arberet, 2008, Robust estimation and blind learning of models for audio source separation.
  
• Alexey Ozerov, 2006, Statistical model adaptation for single-channel source separation. Application to the separation of voice from music in popular songs.
  
• Sylvain Lesage, 2006, Adaptive representations for signal description, separation and classification.
  
  

PAST NEWS

• SPARS'09 (Workshop on Signal Processing with Adaptive Sparse/Structured Representations) was held in Saint-Malo, France, April 6-9 2009.
• SPARS'05 (Workshop on Signal Processing with Adaptive Sparse/Structured Representations) was held at IRISA, Rennes, France, 16-18 November 2005.
• A special issue on Sparse Approximations in Signal and Image Processing was published in EURASIP Signal Processing Journal in July 2006 (the call for paper was here)

PAST RESEARCH PROJECTS

• HASSIP EU network Harmonic Analysis and Statistics in Signal and Image Processing: Mailing list, HASSIP local page
• MathSTIC CNRS project Sparse and structured approximations for audio signal processing
  
• GDR ISIS workgroup on resources for audio source separation:
  

  Discussion list
  Workgroup home page
  ADTS group at IRCCyN
  Analysis-synthesis group at IRCAM
  

SOME ADDITIONAL COLLABORATIONS